This invention relates to scanning probe microscopy, and, more particularly, to an instrument for the simultaneous acquisition of electrical and topographical information about a surface under electrochemical potential control.
This invention was made with government support under Contract No. BIR-9513233 awarded by the National Science Foundation. The government has certain rights in the invention.
The scanning electrochemical microscope (SECM) is a device for measuring the currents owing to electrochemically active species at, or near, a surface, and for mapping their distribution with a spatial resolution on the order of micrometers. FIG. 1 shows the schematic layout of a SECM as taught by Kwak et al, U.S. Pat. No. 5,202,004 and Bard et al, "Scanning Electrochemical Microscopy," Electroanalytical Chemistry, vol. 18:243-373 (1993). The microscope includes a small metal electrode 1, made from a wire 2 covered by insulation 3 which is cut away at one end to expose the inner conductor. The electrode is placed in a solvent 4 containing dissolved ions 9, 10. The tip of the electrode is held in place some distance above a sample to be examined 5 which may, or may not be a conductor. The electrode 1, is connected to a potentiostat 8, to which is also connected an auxiliary electrode 7 and a reference electrode 6. Sample 5, if conducting, may also be connected to the potentiostat. The potentiostat is used to control the potential of the electrode 1 and the sample 5 (if conductive) with respect to the reference electrode 6 by means of a potential applied to the auxiliary electrode, as is well known to those skilled in the art.
The dissolved ions 9, 10 can exist in one of several charge states, for example Fe.sup.++ or Fe.sup.+++. Referring to the less positively charged state as R and the more positively charged state as O, these ions, together with their associated dissolved anions, form a mediator, so called because they mediate the currents that flow between the electrodes. Suitable salts for forming mediator solutions are described in Bard et al. The two charged species exist in equilibrium at an electrode held at the formal potential, E.sub.0 for the process O{character pullout}R.
If the electrode 1 is held negative of the formal potential, species O become reduced to R, giving rise to a current flow through the electrode 1. As a result, the concentration of species O falls in the vicinity of the electrode 1, so that the current also falls. Eventually, the current falls to an equilibrium value determined by the geometry of the electrode and the speed with which replacement ions O can diffuse to the electrode 1. For a disk electrode of radius a, this limiting current is given by the equation: EQU I.sub.L =4nFDc (1)
where n is the number of electrons transferred at each reduction, F is the Faraday constant (9.6.times.10.sup.4 Coulombs per mole of charge), D is the diffusion constant of the ions (often assumed to be the same for O as R and on the order of 5.times.10.sup.-6 cm.sup.2 /s) and c is the concentration (in moles per cm.sup.3, if a is in cm and D is in cm.sup.2 /s). The time for the equilibrium current to be reached is small in the case of a small electrode, being on the order of a.sup.2 /D, or only a few milliseconds where a is on the order of a micron and D=5.times.10.sup.-6 cm.sup.2 /s.
The SECM profiles a surface by utilizing the manner in which the surface affects the diffusion of ions to the electrode. If, for example, the sample surface 5 is an insulator, it blocks the flow of ions to the electrode if the electrode is placed within a distance on the order of its diameter (d=2a) of the surface, as illustrated in FIG. 2. The ion species O, 9, is now constrained to flow in from the sides only, flow from below being blocked by the surface 5. If the electrode tip 1 is now scanned over a surface of varying height, then the flow of current will increase as the surface retreats from the tip, and increase if the surface approaches the tip. This current signal may be used to control the position of the tip and to form a map of the surface, as described by Kwak et al, above.
The SECM may also be used to profile conducting surfaces as illustrated in FIG. 3. In this case, the flow of current is enhanced as the surface of sample 5 is approached. This is because ions that are reduced at the electrode 1 may be rapidly re-oxidized at the sample surface, thereby increasing the supply of ions O in the vicinity of the electrode 1.
This scheme suffers two drawbacks: First, the resolution is limited by the exposed electrode area (being about an electrode diameter, d, under optimal circumstances). Second, it is difficult to profile heterogeneous surfaces which consist of both insulating and conducting portions. This is because a conducting surface which recedes from the probe gives a falling current, just as an insulating surface which approaches the tip.
One solution to this problem has been proposed by Bard and Wipf, U.S. Pat. No. 5,382,336, which solution is illustrated in FIG. 4. In this scheme, the electrode 1 is oscillated up and down by an amount .delta. (16) so that the gap d (17) changes from d+.delta. to d-.delta. at the extreme of each oscillation as shown in the FIG. 5 plot of distance versus time. The corresponding oscillating component of the current (i.sub.cond) a conducting surface versus time is shown in FIG. 6. A similar plot of the current (i.sub.ins) for an insulating surface versus time is shown in FIG. 7. The signal for the case of an insulating surface is in phase with the applied modulation, and, consequently, the output of a lock-in detector fed with this signal would be a positive voltage proportional to the amplitude of the oscillating current. The signal for a conducting surface is out of phase with the modulation, and so the output of a lock-in detector fed with this signal would be a negative voltage proportional to the amplitude of the oscillating current signal. In this way, the output of the lock-in detector can be used to generate a feedback signal which has the correct sign in all cases. However, this scheme suffers from the limited resolution inherent in SECM probes with micrometer dimensions.
Attaching the SECM electrode to the force sensing cantilever of an atomic force microscope (AFM) would improve resolution because the high topographical resolution of the AFM could be combined with the chemical sensitivity of the SECM. Macpherson et al, 118 J. Am. Chem. Soc. 6445-52 (1996) have attempted to do this by insulating a conducting AFM probe as illustrated in FIG. 8. An AFM probe 31 is coated on one side with a platinum film 32 contacted by a conducting clip 34. The clip is in turn connected to a conducting wire 36. The entire assembly is coated in a polystyrene film 33 to render it insulating. Operation of the cantilever in an AFM is assumed to have abraded away the insulating film in a small region near the tip 35, leaving an otherwise insulating film on the cantilever. The cantilever is inserted into an electrolyte 4 above a sample 5. Reference 37 and auxiliary electrodes 38 were also inserted into the electrolyte. In this case, the AFM was used for high resolution imaging, and the AFM cantilever coating was used as an electrode to generate a high concentration of the desired ions in the vicinity of the sample 5. SECM imaging was not attempted. This scheme has the drawback that the desired level of insulation is very hard to achieve. The currents through the cantilever are on the order of ten microamperes for an electrolyte concentration of 0.05 mole/liter. Using D=1.3.times.10.sup.-5 cm.sup.2 /sec and I=10.sup.-5 A gives, from equation 1 above, a=0.08 cm, or d on the order of 1 mm. This is a very large exposed electrode area.
None of the existing SECM or AFM prior art techniques can detect the very small currents associated with electrochemical processes in single molecules. Such small currents have been detected by using a well insulated scanning tunneling microscope tip (Fan and Bard, 267 Science 871-74 (1995)) or by working in an insulating fluid (Han, Durantini et al, 101 J. Phys. Chem. 10719-725 (1997)) (where the quantitative advantage of potential control is lost). These experiments show that detection of electrochemical signals from single molecules requires a sensitivity in the picoampere (pA) range, six orders of magnitude smaller than the leakage signal from poorly insulated AFM tips.
Accordingly, there remains a need in the art for a technique and system to produce an SECM signal that is highly localized to a region close to an atomic force microscope tip. There also remains a need for a technique which is able to detect very small currents associated with electrochemical processes which avoids the problems of current leakage from AFM tips. Finally, there also remains a need for a technique and system which is able to acquire simultaneously AFM topographical images and SECM current data.